CN109872011A - Livestock physiological status prediction technique and system based on multivariate logistic regression model - Google Patents
Livestock physiological status prediction technique and system based on multivariate logistic regression model Download PDFInfo
- Publication number
- CN109872011A CN109872011A CN201910202056.8A CN201910202056A CN109872011A CN 109872011 A CN109872011 A CN 109872011A CN 201910202056 A CN201910202056 A CN 201910202056A CN 109872011 A CN109872011 A CN 109872011A
- Authority
- CN
- China
- Prior art keywords
- data
- domestic animal
- logistic regression
- model
- multivariate logistic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The present invention provides a kind of livestock physiological status prediction techniques and system based on multivariate logistic regression model, the described method includes: with the physiologic information and environmental information of sensor collection domestic animal, related data pretreatment to being collected into, it is trained using cross-validation method and grid search and obtains multivariate logistic regression model, using optimal multivariate logistic regression model, animal physiology state is predicted according to the non-real time physiological data of domestic animal；The present invention can carry out the judgement of timeliness to the physiological conditions of domestic animal, and prediction domestic animal is normally in common disease at this time, remind raiser according to the judging result made, and enable users to the abnormal conditions for finding domestic animal in time.
Description
Technical field
The present invention relates to machine learning field, in particular to it is a kind of using multivariate logistic regression to livestock physiology shape
The method and system that state is predicted.
Background technique
Pork is always a kind of China's even favorite meat in the whole world, is shown according to related data, the meat of Chinese
Once having 80% in consumption is pork.Even if by 2014,56,710,000 tons of national pork yield, more than meat total output
65%, more 4 times out are added up than cattle and sheep meat production.The greatest problem faced in the breeding process is exactly the physiology of domestic animal
State, in breeding process probably due to some carelessness result in not quenchable epidemic situation.Infectious disease is a kind of more typical disease
Disease, to the very harmful of domestic animal, if the bad collapse for be easy to causeing entire farm of prevention and control.Swine disease has swine fever, pseudo- mad dog, mouth
Fever aphthous, Haemophilus parasuis, Streptococcus suis, contagious pleuropneumonia, pig globidiosis, eperythrozoonosis, toxoplasm
Disease.There are many just belong to Amphixenosis, such as aftosa, rabies, Japanese Type-B encephalitis, streptococcosis among these.
The prevention and treatment of the disease of domestic animals is mainly taken with the strategy of " putting prevention first and combining prevention with control " at present.Major measure includes
Reinforce feeding management, do environmental sanitation well, reinforce quarantine, immunity inoculation, disinfection and sterilization kills mouse.The purpose of these measures is
The morbidity source for preventing pig, prevents infectious disease on animal infection, reduces the incidence probability of domestic animal.Or when discovery has few part man
Poultry carries out isolation processing to this part immediately after there is infectious symptom, and inoculates to other domestic animals.General
Circulation method can not entirely prevent the intrusion of infectious disease when being prevented and cost is very big in terms of prevention, once there is epidemic disease
Because cannot result in greater loss in first time discovery after feelings.So monitoring the physiological conditions of domestic animal and judgement at any time
The physiological status of domestic animal at this time is particularly useful for healthy aquaculture.
Summary of the invention
In order to solve the problems, such as cannot to know animal physiology state in time in conventional methods where, the invention proposes one kind
Livestock physiological status prediction technique and system based on multivariate logistic regression model.The present invention can be according to the real-time of domestic animal
Physiologic information judges that the physiological characteristic of domestic animal, prediction domestic animal is normally to be under common disorder at this time.
According to an aspect of the present invention, it is pre- to provide a kind of livestock physiological status based on multivariate logistic regression model
Survey method, comprising:
Step 1, with sensor collection animal physiology information and environmental information；
Step 2, the data collected are pre-processed；
Step 3, using pretreated data, multivariate logistic regression is trained using cross-validation method and grid search
Model；
Step 4, using optimal multivariate logistic regression model, according to the real-time physiological of domestic animal and environmental data to family dirty swine
Reason state is predicted.
Further, with sensor collection animal physiology information and environmental information, comprising: handle and analyze using sensor
Hardware detection arrives and by the data of server transport, converts one for physiology and environmental characteristic information that each boss raises
Record, the physiology and environmental characteristic data of every boss poultry are a sample；Structured message is extracted, is separately extracted from record
Domestic animal is normal, the physiologic information under generic disease state and environmental information.
Further, the data collected are pre-processed, comprising:
The standardization MinMaxScaler processing of minimax normalizing is carried out to continuous type attribute；
Solely hot one-hot processing is carried out to Nominal Attribute；
Z-score z-score encoding operation is carried out to the numerical attribute after discrete codes；
The physiological status of domestic animal is encoded, normal condition is encoded to 0, and sick state is encoded according to the different state of an illness
For 1,2 ..., N, using coding result as the label of model training.
Further, multivariate logistic regression model is trained using cross-validation method and grid search, comprising:
The training set of selected cross validation and the ratio broken number of test set, broken number L are 10, the ratio of training set and test set
Example is 9:1, training set is used for model training, test set is used for model testing；
Setting sample number be n, sample characteristics number be m, classification number be N+1, including 1 normal condition classification and it is N number of not
With the sick status categories of the state of an illness；The sample that preprocessed data is arranged is (x, y), wherein x ∈ Rn*mIt is characterized information, y ∈ RnIt is
Supervision message；The value set of y is { 0,1,2 ..., N }, and solution multivariate logistic regression model obtains sample x and belongs to a different category
The probability of k takes the classification of wherein maximum probability as the generic of sample x；Wherein multivariate logistic regression model is:
Wherein, wk∈RmIt is model parameter, indicates the regression coefficient of kth class；Exp () is exponential function；
The parameter that model is adjusted by grid search, finds the optimized parameter of model.
Further, using optimal multivariate logistic regression model, according to the non-real time physiological data of domestic animal to animal physiology state
It is predicted, comprising:
According to cross validation and parameter adjust as a result, selecting the optimal models that optimal parameter constitutes；
The real-time physiological for the domestic animal being collected into and environmental characteristic data are pre-processed by step 1 and step 2；
By treated, characteristic is put into optimal models, obtains prediction result, if result is 0, the physiology of domestic animal
State is normal.If result is not 0, the state of an illness physiological status of domestic animal at this time is judged according to resulting result.
According to another aspect of the present invention, a kind of livestock physiological status based on multivariate logistic regression model is provided
Forecasting system, the system comprises:
Data collection module, for utilizing sensor collection animal physiology information and environmental information；
Data preprocessing module, for being pre-processed to the data collected；
Model training module, for being trained multivariate logistic regression model using cross-validation method and grid search；
Physiological status prediction module, for utilizing optimal multivariate logistic regression model, according to the non-real time physiological data of domestic animal
Animal physiology state is predicted.
The invention proposes a kind of livestock physiological status prediction technique and system based on multivariate logistic regression model,
The physiological characteristic of domestic animal can be judged according to the real-time physiological information of domestic animal, prediction domestic animal is normal at this time or place
Under common disorder.The present invention can help raiser to know animal physiology state and problem in time, when solving prevention not
The problem of capable of entirely preventing the intrusion of infectious disease, can reduce the cost in terms of prevention, can be at the first time after there is epidemic situation
It was found that reduce loss.
Detailed description of the invention
Fig. 1 is a kind of livestock physiological status prediction based on multivariate logistic regression model provided in an embodiment of the present invention
Method flow diagram；
Fig. 2 is a kind of livestock physiological status prediction based on multivariate logistic regression model provided in an embodiment of the present invention
The schematic diagram of sensor collection data in method；
Fig. 3 be another embodiment of the present invention provides a kind of livestock physiological status based on multivariate logistic regression model
Prediction technique logical flow chart；
Fig. 4 is a kind of livestock physiological status prediction based on multivariate logistic regression model provided in an embodiment of the present invention
System construction drawing.
Specific embodiment
Specific embodiment of the present invention is illustrated below, to further illustrate starting point of the invention and corresponding
Technical solution.
Fig. 1 is a kind of livestock physiological status prediction based on multivariate logistic regression model provided in an embodiment of the present invention
Method flow diagram, as shown in Figure 1, the method includes following four steps:
Step 1, with sensor collection animal physiology information and environmental information；
Step 2, the data collected are pre-processed；
Step 3, using pretreated data, multivariate logistic regression is trained using cross-validation method and grid search
Model；
Step 4, using optimal multivariate logistic regression model, according to the real-time physiological of domestic animal and environmental data to family dirty swine
Reason state is predicted.
Fig. 2 is a kind of livestock physiological status prediction based on multivariate logistic regression model provided in an embodiment of the present invention
The schematic diagram of sensor collection data in method.Preferably, step 1, believed with sensor collection animal physiology information and environment
Breath, specifically includes:
It handles and analyzes and detected using sensor hardware and by the data of server transport, each boss is raiseeed
Physiology and environmental characteristic information are converted into a record, and the physiology and environmental characteristic data of every boss poultry are a sample；It extracts
Structured message, from separately extracting in record, domestic animal is normal, the physiologic information under generic disease state and environmental information.It is described
Domestic animal include pig, sheep, ox etc..
The sensor includes temperature sensor, humidity sensor, gas sensor, physical trait sensor etc..Wherein
Temperature sensor uses noncontacting proximity sensor, and humidity sensor uses low humidity end sensor, and gas sensor is using solid electricity
Matter gas sensor is solved, physical trait sensor can be capable of " the pig ear of real-time monitoring domestic animal physical characteristic information using one kind
Nail " sensor.
Preferably, step 2 pre-processes the related data collected, enables the algorithm to receive treated data, tool
Body includes:
The standardization MinMaxScaler processing of minimax normalizing is carried out to continuous type attribute；
One-hot processing is carried out to Nominal Attribute；
Z-score encoding operation is carried out to the numerical attribute after discrete codes；
The physiological status of domestic animal is encoded, normal condition is encoded to 0, and sick state is encoded according to the different state of an illness
For 1,2 ..., N, using coding result as the label of model training；
Fig. 3 is another enforceable livestock physiological status prediction side based on multivariate logistic regression model of the invention
The logical flow chart of method.Step 1, the received data of collecting sensor, step 2, pretreatment obtain making the acceptable number of algorithm
According to then step 3 being trained and obtain multivariate logistic regression model, step 4, utilization is optimal using pretreated data
Multivariate logistic regression model predicts animal physiology state.
Preferably, step 3 is trained and obtains multivariate logistic regression model using cross-validation method and grid search,
It specifically includes:
The training set of selected cross validation and the ratio broken number of test set, broken number L are 10, the ratio of training set and test set
Example is 9:1, training set is used for model training, test set is used for model testing；
Setting sample number be n, sample characteristics number be m, classification number be N+1, including 1 normal condition classification and it is N number of not
With the sick status categories of the state of an illness；The sample that preprocessed data is arranged is (x, y), wherein x ∈ Rn*mIt is characterized information, y ∈ RnIt is
Supervision message；The value set of y is { 0,1,2 ..., N }, and solution multivariate logistic regression model obtains sample x and belongs to a different category
The probability of k takes the classification of wherein maximum probability as the generic of sample x；Wherein multivariate logistic regression model is:
Wherein, wk∈RmIt is model parameter, indicates the regression coefficient of kth class；Exp () is exponential function；
The parameter that model is adjusted by grid search, finds the optimized parameter of model.
Preferably, step 4 utilizes optimal multivariate logistic regression model, according to the non-real time physiological data of domestic animal to family dirty swine
Reason state is predicted, is specifically included:
According to cross validation and parameter adjust as a result, selecting the optimal models that the optimal parameter in the inside constitutes；
The real-time physiological characteristic of domestic animal is pre-processed, model is enable to receive；
Will treated that characteristic is put into optimal models, obtain as a result, if 0, then the physiological status of domestic animal is being just
Often.If result is not 0, the state of an illness physiological status of domestic animal at this time is judged according to resulting result.
Fig. 4 is a kind of livestock physiological status prediction based on multivariate logistic regression model provided in an embodiment of the present invention
System construction drawing, the system comprises:
Data collection module 401, for utilizing sensor collection animal physiology information and environmental information；
Data preprocessing module 402, for being pre-processed to the data collected；
Model training module 403, for being trained multivariate logistic regression mould using cross-validation method and grid search
Type；
Physiological status prediction module 404, for utilizing optimal multivariate logistic regression model, according to the real-time physiological of domestic animal
Data predict animal physiology state.
Preferential, data collection module is specific to use for utilizing sensor collection animal physiology information and environmental information
In:
It handles and analyzes and detected using sensor hardware and by the data of server transport, each boss is raiseeed
Physiology and environmental characteristic information are converted into a record, and the physiology and environmental characteristic data of every boss poultry are a sample；It extracts
Structured message, from separately extracting in record, domestic animal is normal, the physiologic information under generic disease state and environmental information.
Preferably, data preprocessing module is specifically used for for pre-processing to the data collected:
The standardization MinMaxScaler processing of minimax normalizing is carried out to continuous type attribute；
Solely hot one-hot processing is carried out to Nominal Attribute；
Z-score z-score encoding operation is carried out to the numerical attribute after discrete codes；
The physiological status of domestic animal is encoded, normal condition is encoded to 0, and sick state is encoded according to the different state of an illness
For 1,2 ..., N, using coding result as the label of model training.
Preferably, model training module, for being trained multivariate logistic regression using cross-validation method and grid search
Model is specifically used for:
The training set of selected cross validation and the ratio broken number of test set, broken number L are 10, the ratio of training set and test set
Example is 9:1, training set is used for model training, test set is used for model testing；
Setting sample number be n, sample characteristics number be m, classification number be N+1, including 1 normal condition classification and it is N number of not
With the sick status categories of the state of an illness；The sample that preprocessed data is arranged is (x, y), wherein x ∈ Rn*mIt is characterized information, y ∈ RnIt is
Supervision message；The value set of y is { 0,1,2 ..., N }, and solution multivariate logistic regression model obtains sample x and belongs to a different category
The probability of k takes the classification of wherein maximum probability as the generic of sample x；Wherein multivariate logistic regression model is:
Wherein, wk∈RmIt is model parameter, indicates the regression coefficient of kth class；Exp () is exponential function；
The parameter that model is adjusted by grid search, finds the optimized parameter of model.
Preferably, physiological status prediction module, for utilizing optimal multivariate logistic regression model, according to the real-time of domestic animal
Physiological data predicts animal physiology state, is specifically used for:
According to cross validation and parameter adjust as a result, selecting the optimal models that optimal parameter constitutes；
By data collection module and data preprocessing module by the real-time physiological for the domestic animal being collected into and environmental characteristic number
According to being pre-processed；
By treated, characteristic is put into optimal models, obtains prediction result, if result is 0, the physiology of domestic animal
State is normal.If result is not 0, the state of an illness physiological status of domestic animal at this time is judged according to resulting result.
It is specific embodiments of the present invention and the technical principle used described in above, if conception under this invention
Made change when the spirit that generated function is still covered without departing from specification and attached drawing, should belong to the present invention
Protection scope.
Claims (10)
1. a kind of livestock physiological status prediction technique based on multivariate logistic regression model, which is characterized in that the method packet
It includes:
Step 1, with sensor collection animal physiology information and environmental information；
Step 2, the data collected are pre-processed；
Step 3, using pretreated data, multivariate logistic regression model is trained using cross-validation method and grid search；
Step 4, using optimal multivariate logistic regression model, according to the real-time physiological of domestic animal and environmental data to animal physiology state
It is predicted.
2. the method according to claim 1, wherein with sensor collection animal physiology information and environmental information,
Include:
Handle and analyze and detected using sensor hardware and by the data of server transport, the physiology that each boss is raiseeed and
Environmental characteristic information is converted into a record, and the physiology and environmental characteristic data of every boss poultry are a sample；Extract structuring
Information, from separately extracting in record, domestic animal is normal, the physiologic information under generic disease state and environmental information.
3. the method according to claim 1, wherein being pre-processed to the data collected, comprising:
The standardization MinMaxScaler processing of minimax normalizing is carried out to continuous type attribute；
Solely hot one-hot processing is carried out to Nominal Attribute；
Z-score z-score encoding operation is carried out to the numerical attribute after discrete codes；
The physiological status of domestic animal is encoded, normal condition is encoded to 0, sick state is encoded to 1 according to the different state of an illness,
2 ..., N, using coding result as the label of model training.
4. the method according to claim 1, wherein being trained using cross-validation method and grid search polynary
Logic Regression Models, comprising:
The training set of selected cross validation and the ratio broken number of test set, broken number L are 10, and the ratio of training set and test set is 9:
1, training set is used for model training, test set is used for model testing；
Setting sample number is n, and sample characteristics number is m, and classification number is N+1, including 1 normal condition classification and N number of different state of an illness
Sick status categories；The sample that preprocessed data is arranged is (x, y), wherein x ∈ Rn*mIt is characterized information, y ∈ RnIt is supervision letter
Breath；The value set of y is { 0,1,2 ..., N }, solves multivariate logistic regression model and obtains sample x and belongs to a different category the general of k
Rate takes the classification of wherein maximum probability as the generic of sample x；Wherein multivariate logistic regression model is:
Wherein, wk∈RmIt is model parameter, indicates the regression coefficient of kth class；Exp () is exponential function；
The parameter that model is adjusted by grid search, finds the optimized parameter of model.
5. the method according to claim 1, wherein using optimal multivariate logistic regression model, according to domestic animal
Non-real time physiological data predicts animal physiology state, comprising:
According to cross validation and parameter adjust as a result, selecting the optimal models that optimal parameter constitutes；
The real-time physiological for the domestic animal being collected into and environmental characteristic data are pre-processed by step 1 and step 2；
By treated, characteristic is put into optimal models, obtains prediction result, if result is 0, the physiological status of domestic animal
Normally；If result is not 0, the state of an illness physiological status of domestic animal at this time is judged according to resulting result.
6. a kind of livestock physiological status forecasting system based on multivariate logistic regression model, which is characterized in that the system packet
It includes:
Data collection module, for utilizing sensor collection animal physiology information and environmental information；
Data preprocessing module, for being pre-processed to the data collected；
Model training module, for being trained multivariate logistic regression model using cross-validation method and grid search；
Physiological status prediction module, for utilizing optimal multivariate logistic regression model, according to the non-real time physiological data of domestic animal to family
Poultry physiological status is predicted.
7. system according to claim 6, which is characterized in that data collection module, for utilizing sensor collection domestic animal
Physiologic information and environmental information, are specifically used for:
Handle and analyze and detected using sensor hardware and by the data of server transport, the physiology that each boss is raiseeed and
Environmental characteristic information is converted into a record, and the physiology and environmental characteristic data of every boss poultry are a sample；Extract structuring
Information, from separately extracting in record, domestic animal is normal, the physiologic information under generic disease state and environmental information.
8. system according to claim 6, which is characterized in that data preprocessing module, for the data collected into
Row pretreatment, is specifically used for:
The standardization MinMaxScaler processing of minimax normalizing is carried out to continuous type attribute；
Solely hot one-hot processing is carried out to Nominal Attribute；
Z-score z-score encoding operation is carried out to the numerical attribute after discrete codes；
The physiological status of domestic animal is encoded, normal condition is encoded to 0, sick state is encoded to 1 according to the different state of an illness,
2 ..., N, using coding result as the label of model training.
9. system according to claim 6, which is characterized in that model training module, for using cross-validation method and net
Lattice search is trained multivariate logistic regression model, is specifically used for:
The training set of selected cross validation and the ratio broken number of test set, broken number L are 10, and the ratio of training set and test set is 9:
1, training set is used for model training, test set is used for model testing；
Setting sample number is n, and sample characteristics number is m, and classification number is N+1, including 1 normal condition classification and N number of different state of an illness
Sick status categories；The sample that preprocessed data is arranged is (x, y), wherein x ∈ Rn*mIt is characterized information, y ∈ RnIt is supervision letter
Breath；The value set of y is { 0,1,2 ..., N }, solves multivariate logistic regression model and obtains sample x and belongs to a different category the general of k
Rate takes the classification of wherein maximum probability as the generic of sample x；Wherein multivariate logistic regression model is:
Wherein, wk∈RmIt is model parameter, indicates the regression coefficient of kth class；Exp () is exponential function；
The parameter that model is adjusted by grid search, finds the optimized parameter of model.
10. system according to claim 6, which is characterized in that physiological status prediction module, for polynary being patrolled using optimal
Regression model is collected, animal physiology state is predicted according to the non-real time physiological data of domestic animal, is specifically used for:
According to cross validation and parameter adjust as a result, selecting the optimal models that optimal parameter constitutes；
By data collection module and data preprocessing module by the real-time physiological for the domestic animal being collected into and environmental characteristic data into
Row pretreatment；
By treated, characteristic is put into optimal models, obtains prediction result, if result is 0, the physiological status of domestic animal
Normally；If result is not 0, the state of an illness physiological status of domestic animal at this time is judged according to resulting result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910202056.8A CN109872011A (en) | 2019-03-18 | 2019-03-18 | Livestock physiological status prediction technique and system based on multivariate logistic regression model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910202056.8A CN109872011A (en) | 2019-03-18 | 2019-03-18 | Livestock physiological status prediction technique and system based on multivariate logistic regression model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109872011A true CN109872011A (en) | 2019-06-11 |
Family
ID=66920723
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910202056.8A Pending CN109872011A (en) | 2019-03-18 | 2019-03-18 | Livestock physiological status prediction technique and system based on multivariate logistic regression model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109872011A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113488175A (en) * | 2021-09-07 | 2021-10-08 | 东莞中融数字科技有限公司 | Livestock health monitoring method and device, storage medium and electronic equipment |
| CN115211388A (en) * | 2021-04-15 | 2022-10-21 | 深圳市中融数字科技有限公司 | Livestock health monitoring method and device, storage medium and electronic equipment |
| WO2023123913A1 (en) * | 2021-12-31 | 2023-07-06 | 深圳云天励飞技术股份有限公司 | Medical condition test model training method and apparatus, test method and apparatus, and electronic device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488889A (en) * | 2013-09-18 | 2014-01-01 | 刘华锋 | Method and system of detecting artifacts in ICU patient records based on multivariate logistic regression |
| CN104239554A (en) * | 2014-09-24 | 2014-12-24 | 南开大学 | Cross-domain and cross-category news commentary emotion prediction method |
| CN104936439A (en) * | 2012-12-02 | 2015-09-23 | 安格瑞卡姆有限公司 | Systems and methods for predicting the outcome of a state of a subject |
| CN106886782A (en) * | 2017-01-16 | 2017-06-23 | 浙江大学 | The stratification complexity activity recognition method of fusional movement and physiology sensing data |
-
2019
- 2019-03-18 CN CN201910202056.8A patent/CN109872011A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104936439A (en) * | 2012-12-02 | 2015-09-23 | 安格瑞卡姆有限公司 | Systems and methods for predicting the outcome of a state of a subject |
| CN103488889A (en) * | 2013-09-18 | 2014-01-01 | 刘华锋 | Method and system of detecting artifacts in ICU patient records based on multivariate logistic regression |
| CN104239554A (en) * | 2014-09-24 | 2014-12-24 | 南开大学 | Cross-domain and cross-category news commentary emotion prediction method |
| CN106886782A (en) * | 2017-01-16 | 2017-06-23 | 浙江大学 | The stratification complexity activity recognition method of fusional movement and physiology sensing data |
Non-Patent Citations (1)
| Title |
|---|
| WEI ZHAO等: "Support vector machines classifiers of physical activities in preschoolers", 《THE PHYSIOLOGICAL SOCIETY》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115211388A (en) * | 2021-04-15 | 2022-10-21 | 深圳市中融数字科技有限公司 | Livestock health monitoring method and device, storage medium and electronic equipment |
| CN115211388B (en) * | 2021-04-15 | 2023-12-05 | 深圳市中融数字科技有限公司 | Livestock health monitoring method and device, storage medium and electronic equipment |
| CN113488175A (en) * | 2021-09-07 | 2021-10-08 | 东莞中融数字科技有限公司 | Livestock health monitoring method and device, storage medium and electronic equipment |
| WO2023123913A1 (en) * | 2021-12-31 | 2023-07-06 | 深圳云天励飞技术股份有限公司 | Medical condition test model training method and apparatus, test method and apparatus, and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bao et al. | Artificial intelligence in animal farming: A systematic literature review | |
| CN109872011A (en) | Livestock physiological status prediction technique and system based on multivariate logistic regression model | |
| CN103488148B (en) | A kind of animal behavior intelligent monitor system based on Internet of Things and computer vision | |
| CN103475736B (en) | Sensor communication system and method for conducting monitoring through same | |
| CN109060143B (en) | Pig body temperature monitoring system and method based on thermal infrared technology | |
| AU2010219406A1 (en) | Image analysis for making animal measurements | |
| CN107692980A (en) | A kind of automatic method for monitoring and analyzing of animal health situation and system | |
| Brown-Brandl et al. | A review of passive radio frequency identification systems for animal monitoring in livestock facilities | |
| CN108207700B (en) | Cow information monitoring method and system | |
| CN106037685A (en) | Intelligent analysis system for sow oestrous diagnosis | |
| WO2021104007A1 (en) | Method and device for animal state monitoring, electronic device, and storage medium | |
| KR102315991B1 (en) | Device and system for managing livestck remotly using artificial intelligence | |
| KR102265809B1 (en) | Method and apparatus for detecting behavior pattern of livestock using acceleration sensor | |
| Tian et al. | Real-time behavioral recognition in dairy cows based on geomagnetism and acceleration information | |
| CN106455525A (en) | System and method for pet behavioral identification | |
| Lee et al. | Analysis of growth performance in swine based on machine learning | |
| Syarif et al. | Health monitoring and early diseases detection on dairy cow based on internet of things and intelligent system | |
| CN210746677U (en) | Health monitoring system for livestock animals | |
| CN108805736A (en) | Cultivating system and state monitoring method under two dot patterns of one kind | |
| CN117473446B (en) | Anomaly detection and processing method and system based on deep learning | |
| KR20210067602A (en) | Method for Establishing Prevention Boundary of Epidemics of Livestock Based On Image Information Analysis | |
| CN112101070A (en) | Animal identity recognition system and method for improving recognition rate by nasal print | |
| Witte et al. | Evaluation of deep learning instance segmentation models for pig precision livestock farming | |
| KR102527058B1 (en) | Apparatus for detecting mounting behavior of cattle | |
| CN110123484A (en) | A kind of livestock childbirth detection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190611 |
|
| RJ01 | Rejection of invention patent application after publication |